plot_dimred: Plot a trajectory in a (given) dimensionality reduction

Description Usage Arguments Value Examples

Description

Plot a trajectory in a (given) dimensionality reduction

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
plot_dimred(
  trajectory,
  color_cells = c("auto", "none", "grouping", "feature", "milestone", "pseudotime"),
  dimred = ifelse(dynwrap::is_wrapper_with_dimred(trajectory), NA,
    dyndimred::dimred_landmark_mds),
  plot_trajectory = dynwrap::is_wrapper_with_trajectory(trajectory) &&
    !plot_milestone_network,
  plot_milestone_network = FALSE,
  label_milestones = dynwrap::is_wrapper_with_milestone_labelling(trajectory),
  alpha_cells = 1,
  size_cells = 2.5,
  border_radius_percentage = 0.1,
  size_milestones = 6,
  size_transitions = 2,
  hex_cells = ifelse(length(trajectory$cell_ids) > 10000, 100, FALSE),
  grouping = NULL,
  groups = NULL,
  feature_oi = NULL,
  color_milestones = c("auto", "given", "cubeHelix", "Set3", "rainbow"),
  milestones = NULL,
  milestone_percentages = NULL,
  pseudotime = NULL,
  expression_source = "expression",
  arrow = grid::arrow(type = "closed", length = unit(0.1, "inches")),
  color_density = c("none", "grouping", "feature"),
  padding = 0.1,
  nbins = 1000,
  bw = 0.2,
  density_cutoff = 0.3,
  density_cutoff_label = density_cutoff/10,
  waypoints = dynwrap::select_waypoints(trajectory),
  trajectory_projection_sd = sum(trajectory$milestone_network$length) * 0.05,
  color_trajectory = "none"
)

Arguments

trajectory

A dynwrap trajectory.

color_cells

How to color the cells.

  • "auto": Try to figure out how to color cells depending on whether one of the grouping, feature_io, milestones or pseudotime parameters are defined.

  • "none": Cells are not coloured.

  • "grouping": Cells are coloured according to a grouping (e.g. clustering). Either the grouping parameter or trajectory$grouping must be a named character vector.

  • "feature": Cells are coloured according to the values of a given feature (e.g. gene expression). Either the expression_source parameter or get_expression(trajectory) must be a matrix. Parameter feature_oi must also be defined.

  • "milestone" (recommended): Cells are coloured according their position in the trajectory. The positioning of the cells are determined by parameter milestone_percentages or else by trajectory$milestone_percentages. The colours of the milestones can be determined automatically or can be specified by passing a tibble containing character columns milestone_id and color (See add_milestone_coloring() for help in constructing this object).

  • "pseudotime": Cells are coloured according to the pseudotime value from the root.

dimred

Can be

  • A function which will perform the dimensionality reduction, see dyndimred::list_dimred_methods()

  • A matrix with the dimensionality reduction, with cells in rows and dimensions (comp_1, comp_2, ...) in columns

plot_trajectory

Whether to plot the projected trajectory on the dimensionality reduction

plot_milestone_network

Whether to plot the projected milestone network on the dimensionality reduction

label_milestones

How to label the milestones. Can be TRUE (in which case the labels within the trajectory will be used), "all" (in which case both given labels and milestone_ids will be used), a named character vector, or FALSE

alpha_cells

The alpha of the cells

size_cells

The size of the cells

border_radius_percentage

The fraction of the radius that is used for the border

size_milestones

The size of the milestones

size_transitions

The size of the trajectory segments

hex_cells

The number of hexes to use, to avoid overplotting points. Default is FALSE if number of cells <= 10000.

grouping

A grouping of the cells (e.g. clustering) as a named character vector.

groups

A tibble containing character columns group_id and color. If NULL, this object is inferred from the grouping itself.

feature_oi

The name of a feature to use for colouring the cells.

color_milestones

Which palette to use for colouring the milestones

  • auto: Determine colours automatically. If color is already specified in milestones tibble, this will be used. Otherwise, the colour scheme is determined by milestone_palette_list$auto.

  • given: The milestones object already contains a column color.

  • cubeHelix: Use the rje::cubeHelix() palette.

  • Set3: Use the RColorBrewer::brewer.pal(name = "Set3") palette.

  • rainbow: Use the grDevices::rainbow() palette.

milestones

Tibble containing the column milestone_id (character). If color_milestones is set to "given", this tibble should also contain a column color (character), containing colour hex codes (e.g. "#123456").

milestone_percentages

The milestone percentages.

pseudotime

The pseudotime from the root of the trajectory to the cells as a named numeric vector.

expression_source

Source of the expression

arrow

The type and size of arrow in case of directed trajectories. Set to NULL to remove arrow altogether.

color_density

How to color density, can be "none", "grouping", or "feature".

padding

The padding in the edges to the plot, relative to the size of the plot.

nbins

Number of bins for calculating the density.

bw

Bandwidth, relative to the size of the plot.

density_cutoff

Cutoff for density, the lower the larger the areas.

density_cutoff_label

Cutoff for density for labeling, the lower the further way from cells.

waypoints

The waypoints to use for projecting. Can by generated using dynwrap::select_waypoints().

trajectory_projection_sd

The standard deviation of the Gaussian kernel to be used for projecting the trajectory. This is in the order of magnitude as the lengths of the milestone_network. The lower, the more closely the trajectory will follow the cells.

color_trajectory

How to color the trajectory, can be "nearest" for coloring to nearest cell, or "none".

Value

A dimensionality reduction ggplot of the data.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
data(example_bifurcating)
plot_dimred(example_bifurcating)


# plotting with umap
if (requireNamespace("uwot", quietly = TRUE)) {
  plot_dimred(example_bifurcating, dimred = dyndimred::dimred_umap)
}

# using a custom dimred
dimred <- dyndimred::dimred_mds(example_bifurcating$expression)
plot_dimred(example_bifurcating, dimred = dimred)

# coloring cells by pseudotime
plot_dimred(example_bifurcating, color_cells = "pseudotime")

# coloring cells by cluster
plot_dimred(
  example_bifurcating,
  color_density = "grouping",
  grouping = dynwrap::group_onto_nearest_milestones(example_bifurcating)
)

dynplot documentation built on Dec. 11, 2021, 9:33 a.m.